Kev999

DonM
...Many years ago in this forum I had a detailed posts with screen captures and the workflow detailed out.  I couldn't find it...

 
http://forum.cakewalk.com/tm.aspx?m=1418967
Was it this one?

OMG Ten Years Ago ...
Cool, I searched and couldn't find it - It's been a long time since I've been a regular around here, - I'll have to learn how to search more better more better more :)

An even more foolproof way is to export exactly one bus (whatever is your master). If you choose entire mix and for some reason you have multiple buses or tracks assigned to hardware outs, the mix will include everything including stuff routed to hardware outs that are potentially not heard (because they are muted externally).
Something like that could cause unwanted phase cancellation.
 
If you have exactly one destination bus for the audio the mix should sound identical to what SONAR is outputting. You can verify this by reimporting the mix to a new track and inverting the phase on that track. Playing the project should typically yield complete silence if its an exact copy.

Hey guys, I'm sorry it took so long to respond to your comments. I've been real busy on a non-music related project these past couple of weeks and haven't had a chance to sit down with Sonar again until today. So I reread all your comments and then started to see if I could apply them to my song file. First thing I noticed, right off the bat, is I couldn't find anything called a Master Bus. I did find a Bus A and a Bus B. So I tried Bus A. It seemed to work OK, so I selected Bus A as TTS-1's output and then bounced my tracks through it for the mixdown.
 
I suspect I would find a Master bus if I were trying to mixdown audio tracks. But I'm not -- I'm mixing down MIDI tracks.
 
Sound fidelity wise, it didn't make any difference. The audio file that I listen to inside of Sonar is a substantially fuller sounding file than the audio file that is exported -- or the one that is recorded to a .wav file when I make a .cwb project (yeah, I hunted down the .wav file and found it). It almost sounds as if I'm playing the tune through an AM radio -- not quite that bad though.
 
I think I need to re-emphasize something here. My original tracks are MIDI. There is NO audio. In this one song I'm working on as an example, I'm using Cakewalk's TTS-1 synth to produce audio sounds, which I had originally been sending to my Interface but then I tried Bus A (since I couldn't find a Master bus), and I'm sending the TTS-1 tracks through the conversion process, I guess you'd say. (highlighted all the MIDI tracks and the TTS-1 tracks and bounced these to audio). For the Source Category I've selected Entire Mix, and for Preset, I've left it blank. All other settings I've left alone.
 
I have Sonar configured for 44.1k @ 16 bits depth. That shouldn't matter, should it?
 
I hope this makes things a bit clearer.
 
The last time I was really into doing all this was way back when I was using Pro Audio 9 on a Win98 platform. I don't recall doing anything like the above. I think what I was doing back then was running my PA9's MIDI's audio output to my Mackie 12 channel along with the audio outs from my other MIDI instruments and I mixed things down there, then sent the mix back to PA9 for final recording. I think. There were three or was it four? songs that were a mix of MIDI and audio (guitar and/or voice along with the MIDI instruments) on a few of the tunes.
 
I think I may have some of my old files backed up to some CDs somewhere. I had a catastrophic HD crash several years ago, and I think I lost all my PA9 work files. Anyway, if I can find one of the CDs, I can probaby figure out what I was doing back then.

The Master Bus is one you create by adding a bus to your project and naming it Master. The point of the Master Bus is having that one, final bus where everything is going.

All the points still apply, however, without a Master Bus - though the routing of audio to the Master Bus should make trouble shooting easier. Regardless, try pulling your exported audio into an audio track in your otherwise fully MIDI mix. Invert the phase of that track and playback the full mix - if there are differences you'll hear them. If it is silence, your export is accurately representing the audio.

If the export is accurate, then the points about the playback should be explored. If your playback is through the same audio ports that SONAR is outputting to, you should have the same filters, but perhaps if SONAR is outputting through ASIO and Windows Media Player is using the usual DirectX drivers, then that might be where the audio is changing.
 
I will have to play with exporting a MIDI only project and then importing the resulting file and see if the SUM of the MIDI MIX and the inverted imported wav file cancel. Assuming no random functions in the FX, I'd expect everything to match up.

OK, silly question time...
 
It's been a while since I've worked with only MIDI and TTS. I have a synth connected that is my sound source and I record that to stereo audio. The way to save a MIDI + TTS mix is to bounce it to an audio track first, or at least that's the only way I've found that allows me to actually export audio to a file. Is this how you are creating your exported audio?
 
Something to be aware of when you bounce to track and then play everything back is you are adding the TTS playback over top of the audio track unless you SOLO only that new audio track. That WILL sound fuller because the playback of the MIDI will have a slight timing variation from the Bounced Audio. If you SOLO the Audio only track and listen, it will probably sound like your exported file.
 
 

I usually freeze the track in order to render a soft synth. That's just more convenient than a bounce, even though technically it's doing the same thing.

Welp, I found the problem. It's Windows Media Player. Funny it doesn't do this with audio I've recorded with other DAW software -- at least I think it doesn't. I'll have to go back and give a listen now.
 
Anyway, I have another utility, called VLC Media Player that I normally use to watch movies and videos with, but it will playback just audio if I want. So I loaded up one of the "offending" files into VLC and it sounds identical to what it sounds like inside of Sonar. I even was a-b'ing them in quick succession, and could tell no difference, other than a bit of volume difference between the two platforms.
 
I'm glad to know this but I'm a little disturbed that WMP is messing with the audio. There are no specific settings in WMP that I can locate that allow for the adjustment of WMP's audio fidelity.
 
Oh well, VLC it is from now on. WMP has gotten the boot.
 
Incidentally, I can recommend VLC player for your video needs (as well as audio after today). It does a couple of things I like -- first it can play back just about any format of video files, and will play the files back smoothly -- as long as they've been recorded smoothly, and second, VLC Player has a volume control that takes your sound device's volume up to 200% of standard. I've run it up to 200% occasionally -- with no distortion, just more volume.